Results & outcomes

Every implemented work item gets a baseline, then RankOut re-measures its target metric at 7, 14 and 28 days, classifies the outcome, and recommends the next action.

Baselines

Before a change is applied, RankOut captures a baseline for that work item — the current values of the metrics the action targets (Search Console clicks / impressions / CTR / position for the target query, Analytics sessions, PageSpeed scores), scoped to the target URL and keywords where possible.

Measurement windows

After the change is validated, RankOut schedules re-measurements at 7, 14 and 28 days. Each one compares the current values to the baseline and produces a classification:

Classification	Meaning
Improved	The targeted metrics moved favourably beyond noise.
Unchanged	No meaningful movement yet.
Declined	The metrics moved the wrong way — worth investigating.
Insufficient data	Not enough traffic or history to judge.
Pending	The window hasn't elapsed yet.

An outcome card showing the baseline, the current values and the classification with deltas — Each action's outcome is measured against its own baseline, not a period summary.

The next-best-action loop

When a measurement completes, RankOut proposes an optimisation decision — keep and expand, revise the approach, investigate or roll back, or wait for more data — with a rationale. Accepting it creates a new opportunity, which flows back into the pipeline. Dismissing it records the decision.

RankOut is careful about correlation versus causation. "Improved after this change" is what it claims; it does not assert the change caused the improvement.